A mild thermal and acid-catalyzed rearrangement of O-aryl ethers into ortho-hydroxy arenes

Article abstract of DOI:10.1021/ol051943+

(Chemical Equation Presented) An unusual rearrangement of an O-aryl ether to an ortho-hydroxyaryl system was discovered during our studies on the synthesis of diazonamide A. We discuss the exploration of this rearrangement under mild thermal and both Bron

Full text of DOI:10.1021/ol051943+

ORGANIC
LETTERS
2005
Vol. 7, No. 20
4531-4534
A Mild Thermal and Acid-Catalyzed
Rearrangement of O-Aryl Ethers into
ortho-Hydroxy Arenes
Frederick W. Goldberg, Philip Magnus,* and Rachel Turnbull
Department of Chemistry and Biochemistry, UniVersity of Texas at Austin,
Austin, Texas 78712
p.magnus@mail.utexas.edu
Received August 11, 2005
ABSTRACT
An unusual rearrangement of an O-aryl ether to an ortho-hydroxyaryl system was discovered during our studies on the synthesis of diazonamide
A. We discuss the exploration of this rearrangement under mild thermal and both Brønsted and Lewis acid-catalyzed conditions.
The thermodynamic driving force of a large number of
rearrangement processes involves the migration of an alkyl,
acyl, or aryl group from attachment to a heteroatom (most
frequently oxygen or nitrogen) to a carbon atom.1 Two
venerable reactions that fall under this broad classification
are the well-known aromatic Claisen rearrangement2 and the
Fries rearrangement.3 During our studies on the synthesis
of diazonamide,4 we discovered an unusual rearrangement
that involves the mild thermal conversion of an O-aryl ether
into an ortho-hydroxyaryl system. Warming 1 and 2 (ratio
1:1) converted them into 3 in 80% yield, and 4 was not
observed.5 There appears to be no immediate direct prece-
dence for this rearrangement, although there is some analogy
in the O-aryl glycoside to C-aryl glycoside conversion,6 that
is, the Lewis acid/Brønsted acid-catalyzed transformation of
benzyl ethers of phenols into 2-hydroxydiphenylmethane
derivatives,7a-e and it is known that O-trityl derivatives of
phenols rearrange under acid-catalyzed conditions to C-
tritylated products.8 An unpublished thermal rearrangement
of an O-cresyl ether into a C-cresol derivative9 and the
demonstrated antiproliferative activity of 3,3-diphenyl-1,3-
dihydroindol-2-ones10 provided the motivation to pursue the
rearrangement chemistry described in Scheme 1.
To further explore the rearrangement depicted in Scheme
1, we first converted 5 and 8 into the corresponding O-aryl
(7) (a) Dewar, M. J. S.; Puttnam, N. A. J. Chem. Soc. 1959, 4080-
4086, 4086-4090, 4090-4095. (b) de Mayo, P. Molecular Rearrangements,
Part 1 Aromatic Rearrangements; Interscience: New York, 1963; pp 313-
318. (c) Sprung, M. M.; Wallis, E, S. J. Am. Chem. Soc. 1934, 56, 1715-
1720. (d) Tarbell, D. S.; Petropoulos, J. C. J. Am. Chem. Soc. 1952, 74,
244-248. (e) Hart, H.; Elia, A. R. J. Am. Chem. Soc. 1954, 76, 3031-
3032.
(8) (a) Baeyer, A.; Villiger, V. Ber. 1902, 35, 3013-3035. (b) Van
Alphen, J. Recl. TraV. Chim. 1927, 46, 287-292. (c) Busch, M.; Knoll, R.
Ber. 1927, 60, 2243-2257. (d) Schorigin, P. Ber. 1926, 59, 2502-2510;
1927, 60, 2373-2378; 1928, 61, 277-283. (e) Kharasch, M. S.; Reinmuth,
O.; Mayo. F. R. J. Chem. Educ. 1936, 13, 7-12. (f) Iddles, H. A.; Chadwick,
D. H.; Clapp, J. W.; Mart, R. T. J. Am. Chem. Soc. 1942, 64, 2154-2157.
(g) Burton, H.; Cheeseman, G. W. H. J. Chem. Soc. 1953, 832-837. (h)
McKenzie, C. A.; Chuchani, G. J. Org. Chem. 1955, 20, 336-345.
(9) Bould. L. Studies on the synthesis of tetracycline. Ph.D. Thesis.
Imperial College, London, 1968; p 150. The conversion of I into II is
described.
(1) de Mayo, P. Molecular Rearrangements: Parts 1 and 2; Inter-
science: New York, 1963.
(2) Trost, B. M.; Fleming, I. ComprehensiVe Organic Synthesis; Perga-
mon Press: Oxford, 1991; Vol. 5, pp 827-866.
(3) Trost, B. M.; Fleming, I. ComprehensiVe Organic Synthesis; Perga-
mon Press: Oxford, 1991; Vol. 5, pp 745-760.
(4) For the correct structure of diazonamide A, see: (a) Li, J.; Jeong,
S.; Esser, L.; Harran, P. G. Angew. Chem., Int. Ed. 2001, 40, 4765-4770.
(b) Li, J.; Burgett, A. W. G.; Esser, L.; Amezcua, C.; Harran, P. G. Angew.
Chem., Int. Ed. 2001, 40, 4770-4773.
(5) Unpublished results.
(6) (a) Booma, C.; Balasubramanian, K. K. Tetrahedron Lett. 1995, 36,
5807-5810. (b) Pinhey, J. T.; Xuan, P. T. Aust. J. Chem. 1988, 41, 69-
80.
(10) Natarajan, A.; Fan, Y.-H.; Chen, H.; Guo, Y.; Lyasere, J.; Harbinski,
F.; Christ, W. J.; Aktas, H.; Halperin, J. A. J. Med. Chem. 2004, 47, 1882-
1885.
10.1021/ol051943+ CCC: $30.25
© 2005 American Chemical Society
Published on Web 08/27/2005

Scheme 1. Thermal Rearrangement in the Diazonamide Series
Table 1.
15
R1
R2
yield (%) 17
R1
R2
yield (%)
15a
15b Me
15c iPr
15d NO2
15e Cl
15f MeCO
15g OMe
15h
15i
H
H
H
H
H
H
H
H
Me
OMe
73
76
64
79
76
78
n/a
74
n/a
17a
17b Me
17c iPr
17d NO2
17e Cl
17f MeCO
17g OMe
17h
17i
H
H
H
H
H
H
H
H
Me
OMe
75
68
63
73
71
68
50
85
55
H
H
H
H
15g and 15i, only the C-arylated products 17g and 17i were
isolated. When a hindered 2,6-dimethyl phenol was used,
only the C-arylated 16 was observed.
derivatives 6 and 9, respectively, by treatment with Cbz-
Tyr-OtBu/Cs2CO3 (Scheme 2).11 Heating 6 and 9 (ca. 1:1
mixture of diastereomers for each) in a sealed tube at 80 °C
resulted in 7 and 10 (ca. 1:1 mixture of diastereomers for
each), respectively. While the standard spectral information
of the above compounds was in agreement with the proposed
structural changes, we were not able to obtain X-ray
crystallographic data to confirm this. Consequently, the
conversion of 11 into 12 (see X-ray structure in Supporting
Information) and its rearrangement into 13 (see X-ray
structure in Supporting Information) was carried out for
verification of the previous structural assignments (Scheme
2).
Heating 15a-f and 15h in benzene at 80 °C resulted in
clean rearrangement to the 2-hydroxy isomers 17a-f and
17h, respectively. It is noteworthy that the phenol ether 15a
gave only the 2-hydroxy isomer 17a and none of the
4-hydroxy isomer 18a. The next experiments were to
establish whether the rearrangement exhibited crossover.
Subjecting 15b to the same thermal conditions for the
rearrangement but now in the presence of 4-iPrC6H4OH gave
17b and 17c as a 1:1 mixture (Scheme 4).
Scheme 4. Crossover Experiments
Scheme 2. Model Thermal Rearrangements
Clearly, this is classical evidence for the dissociative SN1
mechanism, and as such, should be responsive to both
Brønsted and Lewis acid catalysis.13 This is indeed the case.
A trace of trifluoroacetic acid caused the rearrangement of
15a to 18a to take place at 25 °C within a few hours, and
exposure of 15a to Lewis acids (Cu(OTf)2, AgOTf, TiCl4)
at 25 °C also gave 18a.
At first sight, it seems strange that the thermal dissociation
of 15a should give only ortho-substitution, yet the rear-
rangement exhibits crossover. Consequently, we examined
the acid-catalyzed rearrangement of 15 and performed further
crossover experiments. Treatment of 15a and 15h in benzene
To examine the above rearrangement in more detail, we
treated 1412 with a variety of phenols in CH2Cl2 at 25 °C in
the presence of Cs2CO3, or refluxing CH2Cl2/Et3N in the case
of 15i (Scheme 3, Table 1). In the cases of 15a-f and 15h,
the O-aryl ether was the main product. For the substrates
Scheme 3. Thermal Rearrangements
(11) For references to azaxylylene intermediates formed from 1,4-
elimination, see: (a) Steinhagen, H.; Corey, E. J. Angew. Chem., Int. Ed.
1999, 38, 1928-1931. (b) Fuchs, J. R.; Funk, R. L. J. Am. Chem. Soc.
2004, 126, 5068-5069. (c) Fuchs, J. R.; Funk, R. L. Org. Lett. 2005, 7,
677-680. (d) Avemaria, F.; Vanderheiden, S.; Braese, S. Tetrahedron 2003,
59, 6785-6796.
(12) Sutcliffe, B. J. Chem. Soc. 1957, 4789-4792.
(13) Alder, R. W.; Baker, R.; Brown, J. M. Mechanisms in Organic
Chemistry, DissociatiVe Processess; Wiley-Interscience: New York, 1971;
pp 78-179.
4532
Org. Lett., Vol. 7, No. 20, 2005

at 25 °C with a catalytic amount of CF3CO2H resulted in
the para-substituted rearrangement products 18a and 18b and
no trace (1H NMR and TLC) of 17a or 17b. It was also
noted that treatment of 17h under the same conditions gave
18c presumably via ipso-protonation (Scheme 5).
The key results from Sprung,7c Tarbell,7d and Hart7e
showed that 23 rearranged into 24 with 76% retention of
absolute configuration. Treatment of 25 with p-cresol in H2-
SO4/AcOH gave racemic 24 (Scheme 7). Dewar7a concluded
Scheme 7. Inter- and Intramolecular Rearrangement
Scheme 5. Acid-Catalyzed Rearrangements
from these results and his own studies that “the rearrange-
ment of alkyl aryl ethers to alkyl phenols can take place by
two routes, one intermolecular and one intramolecular”. He
postulated that the reactions proceed via an ortho-π-complex
to explain the predominate formation of 24 rather than its
p-isomer. All of the rearrangements described in the above
literature and in references6,8 are conducted using Lewis acids
or Brønsted acids. There are no reported thermal versions.
Under neutral conditions, the O-aryl ether begins to
dissociate (26) and forms a π-complex 27, which can be
regarded as a cationic azaxylylene intermediate (Scheme 8).
It is noteworthy that the azaxylylene intermediates described
in ref 10 were neutral as they were derived from N-H
precursors. This is the first reported example of the formation
of cationic azaxylylene intermediates derived from N-alkyl
precursors.
The π-complex (27) can rearrange to give 2814 and
subsequently tautomerize to 29. When these reactions are
conducted in the presence of a Brønsted acid, the initial
partially dissociated adduct 26 can completely ionize to give
30 (which is also accessible from 29 through ipso-protonation
to give 28, see the conversion of 17h into 18c), which results
in the thermodynamically more stable p-substituted product
31 (Scheme 8). The original Dewar mechanism provides the
Thermal rearrangement of 15c in the presence of phenol
(5 equiv) gave 17c and 18a (1:4). The ortho-isomer 17a was
not detected. Similarly, CF3CO2H-catalyzed rearrangement
of 15c in the presence of phenol (5 equiv) gave 17c and 18a
(1:5). These results point to two clear mechanistic pathways
for the rearrangement (see later).
It was of some interest to see if the same chemistry
described above was applicable to the NMe series, where
the azaxylylene intermediate would be positively charged
(as in Scheme 8, 27). Treatment of 19 with phenol, p-cresol,
and o-cresol in CH2Cl2 at 25 °C in the presence of Cs2CO3
did not give 20a, 20b, or 20d, respectively. Instead, the
C-arylated products 21a, 21b, and 21d were formed directly
(Scheme 6, Table 2). However, p-nitrophenol and o-nitro-
Scheme 6. NMe Rearrangements
Scheme 8. Proposed Mechanism of Rearrangement
phenol gave 20c and 20e, respectively, which rearranged
upon heating at 80 °C in benzene to give 21c and 21e.
Table 2.
20
R1
R2
yield (%)
21
R1
R2
yield (%)
20a
20b Me
20c
20d
20e
H
H
H
H
Me
n/a
n/a
60
21a
21b Me
21c
21d
21e
H
H
H
H
Me
NO2
63
68
67
57
77
NO2
H
H
NO2
H
H
most plausible explanation and is a reminder of the powerful
effects of ion pairing.15
n/a
NO2 72
(14) Principle of least motion: (a) Rice, F. O.; Teller, E. J. Chem. Phys.
1938, 6, 489-496. (b) Hine, J. AdV. Phys. Org. Chem. 1977, 15, 1-61.
(15) Szwarc, M. Ions and Ion-Pairs in Organic Reactions; Wiley: New
York, 1972.
Exposure of 20e to a catalytic amount of CF3CO2H in CH2-
Cl2 at 25 °C resulted in rearrangement to the p-isomer 22.
Org. Lett., Vol. 7, No. 20, 2005
4533

In conclusion, this subtle and unexpected reaction and its
relationship to the Dewar studies illustrates that natural
product synthesis continues to be one of the central vehicles
for the discovery of new chemistry.
research. Dr. Chi-Ming Cheung is thanked for the conversion
of 8 into 10.
Supporting Information Available: Experimental pro-
cedures and characterization of all new compounds. X-ray
crystallographic data for 12 and 13. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL051943+
Acknowledgment. The NIH (CA RO1 50512) and the
Welch Chair (F-0018) are thanked for their support of this
4534
Org. Lett., Vol. 7, No. 20, 2005